結合文字探勘與量化工具從線上留言挖掘旅館業者的競爭優勢

Combining Text Mining and Quantitative Methods to Explore the Hotels' Competitive Advantage through Online Reviews

林金賢(Chin-Shien Lin) ； 顏素絹(Su-Chuan Yen) ； 徐曉頡(Siao-Jie Syu)

《資訊管理學報》 28卷3期 (2021/07) Pp. 337-360

摘要

線上留言對於了解消費者需求提供一個異於過去的重要管道，也成為旅館業者制定品牌策略的重要資訊來源。對於如何利用線上評論來了解消費者需求並進一步找出各旅館的競爭優勢，在文獻上仍然存在有缺口。本研究主張消費者心中各家旅館的競爭優勢，必須帶入消費者的購買行為，才能被挖掘出來。本研究針對TripAdvisor網站，利用爬蟲程式擷取台中市西區著名的五家星級旅館從2019年9月到11月三個月期間的線上評論各100則作為分析對象，利用文字探勘工具找出消費者在乎的旅館屬性，各旅館在各屬性上的表現，並利用對應分析找出消費者心中各旅館的群組歸屬，最後結合消費者購買行為與邏輯斯迴歸找出各旅館的競爭優勢。本研究不但說明如何從線上評論找出各旅館在消費者心目中的策略群組歸屬外，並找出各旅館在其所屬的策略群組中的競爭優勢，在實務上對品牌定位決策可以提供具體作法之外，在學術上也可以填補文獻上在品牌策略制定上的缺口。

關鍵字

線上評論 ； 文字探勘 ； 對應分析 ； 競爭優勢

並列摘要

Purpose - There's still a gap in the research that needs to be fulfilled about how to utilize the on-line reviews to find the customers' needs and to identify the competitive strategies for each hotel. This study claims that only the customers' actual transaction behaviors would reveal the true competitive advantages of the hotels in customers' minds. Therefore, the objectives of the research are finding the strategic group of the hotels in customers' minds and locating the competitive advantages for each hotel in their strategic group from the on-line reviews. Design/methodology/approach - The main methods used for achieving the research objectives are employing the text-mining analysis to find the hotels' attributes for the customers' preferences and the performance of each hotel's attributes, correspondence analysis to identify the position of the underlying group for each hotel and logistic regression to locate the competitive advantages for each hotel. Findings - The result of the text-mining analysis identifies the first 22 favorite attributes of the hotel in the customers' online reviews while the correspondence analysis indicates the results of the perceptual map of the hotels' branding position in the customers' minds. Moreover, the logistic regression shows one hotel has chosen by the customers because of certain attributes and these attributes are the competitive advantages for this hotel and the results will lead the hotels to enhance their competitive advantages. Research limitations/implications - The limitation of this study contains three parts, the first part would be the small number of the sample group which is 500 luxury hotels located in Xitun District of Taichung in this paper. Therefore, future research is suggested to include more hotels and data. Secondly, the limited variables would lead the study to employ the satisfactory value to represent the empirical validity. Adding the repurchase intention as an additional variable would build up the binary logistic regression to become another empirical validity. Last, six dimensions of the services are used to generate the perceptual map in the study, the future research may use more variables to describe the attributes of the hotels to analyze the pros and cons for the hotels . Practical implications - The practical result of the research proves that the customers' online reviews would locate the perspectives for the customers' needs and provides the hotels to enhance their advantage competitiveness. This would bring more values to the hotels. The data for the study gathered from the TripAdvisor on-line reviews is lack of demographic variables which would limit the hotels making decision accurately. Originality/value-This research not only finds the strategic group of the hotels in customers' minds, but also locates the competitive advantages for each hotel in their strategic group from the on-line reviews. Practically, the result of this study would help the hotel industry to make the decision on the brand positioning effectively and academically, it fills the gap from the past studies on brand positioning while setting up the strategies.

並列關鍵字

Online review ； text mining ； correspondence analysis ； competitive advantage

參考文獻

Aaker, D. A. (1996), ‘Building Strong Brand’, The Free Press, New York.

Alaei, A., Becken, S., and Stantic, B. (2017), ‘Sentiment Analysis in Tourism: Capitalizing on Big Data’, Journal of Travel Research, Vol.58, No.2 , pp. 175-191.

Anderson, R. I., Fish, M., Xia, Y., and Michello, F. (1999), ‘Measuring efficiency in the hotel industry: A stochastic frontier approach’, International journal of hospitality Management, Vol.18, No.1, pp. 45-57.

Ban, H. J. and Jun, J. K. (2019). ‘A study on the semantic network analysis of luxury hotel and business hotel through the big data’, Culinary Science & Hospitality Research, Vol.25, No.1, pp. 18-28.

Brown, J. R., and Ragsdale, C. T. (2002), ‘The competitive market efficiency of hotel brands: An application of data envelopment analysis’, Journal of Hospitality & Tourism Research, Vol.26, No.4, pp. 332-360.